I asked an AI to read all 120 of my notes: 76 retrospectives from 2019 through August 2026 and 44 permanent notes. This is what it found.
Something positive
My strongest quality is adaptive perseverance.
I don’t avoid feeling rejection, uncertainty, burnout, or fear—I feel those things intensely—but I repeatedly turn them into learning and action. I went from wanting someone to give me a chance as a programmer in 2019, through rejection and burnout, to becoming an experienced developer, leading work, speaking publicly, and adapting to AI rather than retreating from it.
My perseverance also isn’t confined to work. It appears in fatherhood, supporting my family, buying and renovating a home, managing health problems, and repeatedly rebuilding routines after they collapse. I sometimes describe myself as inconsistent, but the longer record says otherwise: I am unusually good at restarting.
Something negative, constructively
I take on too many possible versions of myself at once, then interpret the inevitable incomplete progress as a personal failing.
I try to be an exceptional developer, present father and husband, healthy athlete, constant learner, product founder, writer, homeowner, and early adopter—simultaneously. When time cannot accommodate all of that, I often conclude that I lack discipline or am not working hard enough. That creates a recurring loop:
enthusiasm → too many commitments → fragmented progress → self-criticism → anxiety and poor sleep → reduced capacity
I had already identified it by 2020—“spreading myself a little thin”—and again in 2025: “when I try to do everything I do nothing.” In 2026, AI gave me the ability to pursue even more ideas, but I correctly recognized that doing “a million different things” was becoming a recipe for burnout.
The answer is probably not better productivity. It is choosing fewer active identities. I should treat my interests as seasons: one meaningful growth project, a sustainable health minimum, and explicit permission for the rest to wait.
Something I might not know about myself
I don’t think my core identity is “software developer.” I think it is caregiver-builder.
My most meaningful projects repeatedly involve understanding a system well enough to make somebody’s life safer, easier, or more delightful: building Bump Buddy for Jenny, creating my own baby camera, improving planning for colleagues, preserving family photographs, teaching other developers, considering environmentally responsible websites, and evaluating career choices through their effect on my family.
That also changes the diagnosis of my unfinished commercial projects. I’ve called the problem a lack of discipline, but the evidence contradicts that—I have demonstrated enormous discipline when the person, purpose, and finish line are concrete. I struggle more when the goal is abstract, such as “build a profitable product,” and no particular person is waiting for it.
My likely bottleneck isn’t discipline. It’s committing under ambiguity: choosing one ordinary problem, tolerating the possibility that nobody cares, and continuing through distribution and maintenance after the technical novelty disappears.
That is potentially reassuring in the face of AI. Coding is one instrument I use; it isn’t the deepest thing I contribute. My more durable ability is combining care, curiosity, and systems thinking to make useful things for real people. That survives changes in tools—and even a change in career.